In this chapter, the articles included in the study are divided into seven areas according to their application scenarios: trauma and facial reconstruction, orthognathic surgery, implant placement, maxillomandibular surgery, removal of foreign bodies, skull base surgery, and endodontic treatment. The development of CADIN systems in oral and maxillofacial surgery, their advantages in various types of clinical applications, and the current disadvantages they face are described.
3.1 Trauma and Facial Reconstruction
Facial trauma usually refers to facial defects caused by trauma or tumor ablation to the orbital and zygomatic complexes, as well as the maxilla and mandible. Facial reconstruction is one of the most common means of repairing facial trauma. During facial reconstruction, the surgeon first needs to identify the location of the nerves to ensure that no nerve damage occurs during the reconstruction process. Secondly, it is necessary to consider the symmetry of the reconstructed bone with the uninjured side to achieve the facial aesthetic requirements. In addition, maintaining proper orbital projection and restoring orbital volume need to be considered in orbital reconstruction. However, intraoperative visualization of the reconstructed site is limited, which makes maxillofacial trauma reconstruction and aesthetic correction still one of the most difficult and challenging procedures for the surgeon (Sukegawa et al. 2018; Landaeta-Quinones et al. 2018; Louis and Morlandt 2018; Wang and Dillon 2021; Demian et al. 2019). CADIN has become an integral part of maxillofacial reconstruction. Using CADIN, fractures can be repositioned by overlaying preoperatively planned images with intraoperatively acquired images and immediately assessing fracture repositioning. CADIN significantly improves fracture repositioning accuracy and the reconstruction of the orbital volume, reducing operative time. In addition, CADIN can help guide instruments in the surgical field in real time so that surgery can become more precise and minimally invasive (Dubron et al. 2022; Schreurs et al. 2021d).
Due to their prominent location and thin orbital wall, orbital fractures usually occur in the setting of facial trauma, and surgical treatment of orbital fractures focuses on repositioning the orbital contents and the eye and restoring structural support to restore ocular function. In the reconstruction of orbital fractures, precise bone reconstruction and repositioning of the orbital soft tissues will largely alleviate clinical symptoms (Hartmann et al. 2022; Kanno et al. 2019; Sabelis et al. 2022). Guatta and Scolozzi (2018) used a method combining "mirroring" virtual planning and intraoperative navigation guidance to perform bone remodeling in a patient with severe asymmetry of the anterior orbit with fibrous dysplasia. Postoperatively, the patient had good cosmetic and psychological results with no complications (Fig. 4(a)). A retrospective study conducted by Yang and Liao (2019) analyzed the surgical outcomes of 17 patients with complex orbital fractures treated under navigational guidance. Based on the postoperative results, it was seen that the average enophthalmos was reduced from 2.99mm to 0.68mm under intraoperative navigation guidance, and there were no postoperative complications (Fig. 4(b)). Thakker et al. (2019) reported three cases of maxillofacial trauma surgery using virtual planning and intraoperative navigation. Practical clinical applications have shown that intraoperative navigation can be particularly helpful in the repair of delayed enophthalmos (Fig. 4(c)). Nazimi et al. (2019) evaluated the effectiveness of using preoperative planning, intraoperative optical navigation, and intraoperative CT techniques in the reconstructive repair of orbital fractures. After practical clinical application, they concluded that intraoperative CT allows real-time assessment of surgical repositioning of orbit-zygomatic complex fractures and adjustment of intraoperative orbital implant plates for reconstruction of orbital blowout wall fractures. Schreurs et al. (2018) proposed in 2018 an implant-oriented navigation method that provides real-time intuitive feedback during insertion (Fig. 4(d)). In 2020, they compared this real-time navigation method with a marker-based navigation method(Schreurs et al. 2020) . A total of 80 orbital defect reconstructions were performed by two surgeons on 10 cadavers. In the real-time implant-oriented approach, implant positioning was significantly improved in terms of roll, yaw, and translation. Real-time implant-oriented navigation feedback provides surgeons with real-time, intuitive feedback, which improves implant positioning and reduces the duration of the navigation process. Schreurs et al. (2021c) in 2021 also evaluated several techniques for orbital reconstruction and concluded that intraoperative navigation provides more intuitive feedback on implant position, improves implant positioning, and improves implant positioning compared to conventional reconstruction. improved accuracy of implant placement. Real-time navigation proved to be the safest, fastest, and most accurate method for placing orbital implants in the desired position (Fig. 4(e)). Dong et al. (2020) investigated the feasibility of reconstructing isolated medial orbital wall fractures through a transorbital approach using u-HA/PLLA sheet implants with the help of intraoperative navigation. It was found that all patients had improved ophthalmic function and aesthetic outcomes and were treated successfully without any long-term complications during the follow-up period. Zeller et al. (2021) combined a self-centered, second-generation patient-specific functionalized orbital implant with intraoperative navigation in performing orbital reconstructive surgery and compared the accuracy of this approach with that of CAD-based individualized implants. The method they used was more accurate in terms of implant positioning and more conducive to translating virtual planning into actual surgery, according to implant fit analysis (Fig. 4(f)). Parameswaran et al. (2022) investigated whether intraoperative navigation could improve radiographic outcomes in patients who underwent delayed primary/secondary orbital reconstruction for inferomedial defects. A real clinical trial with 40 patients revealed a substantial reduction in intraorbital volume and enophthalmos. Implant positioning using intraoperative navigation was more accurate, with less deviation in mediolateral and yaw, and the operative time for implant positioning was reduced by 17 minutes. Rana et al. (2022) investigated the accuracy of patient-specific implants for selective laser melting combined with intraoperative navigation in primary orbital reconstruction in 96 patients with orbital fractures. The mean volume difference between the planned orbit and the postoperative orbit was less than 1cm3. 3D analysis by color mapping showed that the reconstructed orbit deviated less from the mirror image unaffected side (Fig. 4(g)). Soh et al. (2022) evaluated the accuracy of maxillofacial reconstruction in 112 patients assisted by navigation by overlaying postoperative CT data with a preoperative virtual surgical plan. There was no significant difference in orbital volume or projection between preoperative, postoperative, and healthy orbits (P=0.093 and P=0.225, respectively). The authors noted a significant association between the accuracy of navigation-assisted mandibular reconstruction and the preservation of the condyle, type of reconstruction, type of osteosynthesis plate, and number of bony segments. Navigation-assisted midfacial reconstruction had higher accuracy compared to mandibular reconstruction (Fig. 4(h)). Mixed reality (MR) can address the effects of inadequate visualization of intraoperative 2D images and unintuitive presentation of the affected area and can greatly alleviate the hand-eye incongruity associated with viewing medical images during surgery. Rahimov et al. (2022) incorporated virtual planning and MR techniques into orbital reconstruction surgery, using virtual planning to develop a detailed and accurate surgical plan and then MR to make the preoperatively developed surgical plan visible to the surgeon in real time, providing intraoperative navigation for orbital reconstruction. After the actual clinical trial, it was found that the patient's eye mobility improved and diplopia was reduced after the surgical reconstruction. However, Rahimov et al. also noted that if the surgeon is unable to focus on the holographic and physical objects and lacks experience in gaming technology and augmented reality (AR), it is likely to affect the accuracy of MR as intraoperative navigation (Fig. 4(i)).
The cheekbone is one of the most frequently injured facial bones due to its prominent position. Ideal primary reconstruction is not always possible, especially in traumatic situations, and the complications of secondary reconstruction pose a significant challenge even for experienced surgeons. Traditional intraoperative positioning of the zygoma usually relies on temporary fixation of the zygoma in its new position, followed by a cursory assessment of the repositioning on the operating table. Although the use of this method can reduce operative time, errors and undercorrections are common (Aman et al. 2020). Nguyen et al. (2019) reported a case of revision, repair, and reconstruction of multiple facial fracture malunions using a 3D virtual surgical plan to design the osteotomy and reset the fractured facial bone. Intraoperative navigation was used to guide and confirm the accurate positioning of the bone segments. Postoperative maxillofacial CT revealed excellent positioning of bony segments, improved symmetry, and the establishment of more anatomically correct transverse, vertical, and AP dimensions. Optimal bony reduction is especially evident when comparing the preoperative and postoperative transverse dimensions of the mandible and zygomas (Fig. 4(j)). Buller et al. (2019) compared the results of using ultrasound as intraoperative navigation with conventional palpation in closed reduction surgery for isolated zygomatic arch fractures. In 16 intraoperative ultrasound cases and 60 control cases, postoperative displacement angles were significantly lower in the ultrasound group for all fractures and variable fracture types. The ultrasound group showed an improvement in the overall degree of repositioning in all fractures, but only M-shaped fractures did not differ. Based on the experimental results, they concluded that although reduction control via palpation and probing using the Volkmann’s hook showed satisfactory results for M-shaped fractures, the additional intraoperative ultrasound imaging promises increasing success rates for variable-configured zygomatic arch fractures. Sharma et al. (2021) utilized intraoperative CT for intraoperative evaluation and implant navigation in the treatment of zygomatic-orbital/isolated orbital complex fractures. Intraoperative CT has been shown through practical clinical trials to play an important role in assessing the accuracy of zygomatic-orbital/isolated orbital fracture repositioning and determining implant placement. This can avoid the need for secondary corrective surgery and postoperative imaging. This method is an important tool to improve the surgical outcome of zygomatic-orbital orbital fractures.
The maxilla is an important skeletal structure that maintains the shape of the middle of the face and performs oral functions. Maxillary defects are often accompanied by the destruction or absence of important surrounding structures, resulting in facial deformity and loss of oral functions. Zhang et al. (2021a) performed a maxillary reconstruction with a composite deep circumflex iliac artery flap, assisted by virtual surgical planning and intraoperative navigation. The surgical procedure was completely guided by an intraoperative navigation system. The experimental results suggest that the application of virtual surgical planning and intraoperative navigation has the potential to improve maxillary reconstruction outcomes (Fig. 4(k)).
The mandible forms the bony scaffold of the lower 1/3 of the face and is essential for maintaining facial shape, chewing, and other functions. Mandibular defects are more common among craniofacial defects and can be caused by a variety of diseases, such as tumors and trauma. The resulting mandibular and adjacent soft tissue defects often lead to serious facial deformities and dysfunctions, which seriously affect the physical and mental health of patients. Zheng et al. (2018) evaluated a computer-assisted approach to reconstructing mandibular defects using a vascularized iliac crest flap in 14 patients who underwent mandibular segmental resection. Computer-based surgical techniques, including virtual surgical planning, CAD/CAM, rapid prototyping, and intraoperative navigation, were used. Postoperatively, the 13 patients with successful flaps had no other serious complications, and they demonstrated good mandibular structure and occlusion. In addition, the height of the bone graft was adequate for the implant. Both the recipient and donor sites healed well without serious complications. Zheng et al. concluded that the use of these digital techniques is a viable option for reconstructing mandibular defects. However, they also noted that the number of patients in the cases reported was small and the follow-up time was relatively short, and further validation should be achieved in more patients subsequently. Pietruski et al. (2019a) developed a new technique using intraoperative navigation techniques to support fibula-free flap harvest and fabrication. Although the experiment was performed on a specially manufactured lower leg phantom for a series of navigated fibular osteotomies and was not applied in oral and maxillofacial surgery, Pietruski et al. concluded that the proposed method is particularly suitable for plastic and maxillofacial surgery and may become a routine surgical approach for the reconstruction of complex mandibular defects. Sun et al. (2019) performed transoral mandibular segmental resection followed by vascularized osseous reconstruction using intraoral anastomosis in nine patients using preoperative virtual surgical planning and intraoperative navigation. Postoperative imaging and examination showed that the new mandible was well positioned with good occlusion in all cases, and facial symmetry was achieved in all cases with an unrestricted opening. Sozzi et al. (2022) proposed an innovative protocol for virtual planning and surgical navigation in post-oncological mandibular reconstruction through a fibula free flap and evaluated the accuracy of the surgical protocol. The location of mandibular markers was evaluated for postoperative accuracy in the preoperative and postoperative CT scans of 21 patients who underwent surgical treatment for mandibular tumors. The maximum discrepancy was 3.4mm for undercorrection and 3.2mm for overcorrection. Overall, the accuracy of the protocol proposed by Sozzi et al. is satisfactory and has good generalizability (Fig. 4(l)). Gao et al. (2022a) applied a personalized guide plate combined with intraoperative real-time navigation in 12 cases of gastrocnemius flap repair of mandibular defects to provide a basis for precise mandibular repair and reconstruction. Intraoperatively, the guide plate and navigation were well applied, and the fibula was accurately shaped and positioned. Postoperatively, the fibularis muscle flap survived with good incision healing and a good occlusal relationship. The actual clinical results showed that the personalized guide combined with intraoperative real-time navigation improved the accuracy of mandibular gastrocnemius flap reconstruction, reduced complications, and provided a preliminary basis for the application of intraoperative visual navigation in mandibular gastrocnemius flap reconstruction. In order to achieve intraoperative navigation, the preoperative CT scan needs to be accurately aligned with the patient. During navigation for mandibular reconstruction surgery, the mandible should be kept in a fixed position or its movement should be tracked, as even small movements of the mandible can reduce navigation accuracy. de Geer et al. (2022) developed a noninvasive hybrid registration method for an electromagnetic intraoperative navigation system for mandibular reconstruction surgery. The method uses anatomical markers for initial point alignment and the exposed surface of the mandible for optimized surface alignment. They used 3D printing technology to print three reference notched mandibular models in the osteotomy plane to experimentally validate the registration procedure in a metal-free environment, using dental molds and without dental molds, respectively. In all experiments, the average marker registration error values were less than 2.0mm. No differences were found using different surface point areas or numbers of surface points. The experimental results demonstrate that hybrid registration is a noninvasive method that requires only a small area of the bare surface of the mandible to achieve high accuracy in the model setup.
The condyle is one of the major growth centers of the mandible and also plays an important role in speech, mastication, and other important physiological functions. However, the anatomical structure of the condyle is weakened, and if this area is damaged or destroyed before development is complete, it will affect the growth and development of the mandible, which in turn will lead to maxillofacial deformities. Han et al. (2018) evaluated the effect of surgical navigation on the accuracy of intracondylar capsule fracture repositioning. 20 patients with intracondylar capsule fractures were divided equally into navigation and control groups, and the mean distance between preoperative and postoperative computed tomography measurements was 0.5235mm and 1.170mm for the navigation and control groups, respectively (P<0.001). The anatomical repositioning rates were 93.8% and 88.2% in the navigation and control groups, respectively (P=0.58). The results indicated that the navigation group achieved more accurate repositioning than the control group (Fig. 4(m)).
Craniosynostosis can lead to cranial vault deformity and facial asymmetry, and open cranial vault reconstruction is currently the most common treatment. García-Mato et al. (2019) evaluated a new patient-specific correction workflow for craniotomy based on intraoperative navigation and 3D printing in five patients suffering from craniosynostosis. An intraoperative navigation system based on optical tracking allows for a good transfer of preoperative virtual planning to the intraoperative period through real-time positioning and 3D visualization. Intraoperative surface scanning is used as the gold standard to assess navigation accuracy. The mean errors obtained in the remodeled frontal region and supraorbital bar were 0.62mm and 0.64mm, respectively. Clinical results suggest that intraoperative navigation is an accurate and reproducible technique for the correction of craniosynostosis that optimally transfers the preoperative plan to the intraoperative one (Fig. 4(n, o, p)).
With the development and use of weapons, facial ballistic trauma has become more complex and inhomogeneous. Facial ballistic trauma often implies severe damage to multiple structures of the maxillofacial region, and it has been a challenge to ensure the accuracy of facial reconstruction and restore normal facial function to the patient. Khatib et al. (2018) proposed a protocol for maxillofacial reconstruction after facial ballistic trauma using computer-assisted surgery such as virtual surgical planning and intraoperative navigation techniques to restore orbital volume, facial projection, and facial symmetry in a primary care hospital. Facial projection was satisfactory in all patients. Khatib et al. concluded that the proposed protocol optimized the reconstruction of facial width, projection, and contour to ensure functional and aesthetic appearance after facial reconstruction. Computer-assisted surgery, such as virtual surgical planning and intraoperative navigation techniques, helped to accomplish efficient and accurate primary reconstruction of complex facial ballistic injuries. Kantar et al. (2019) used preoperative diagnostic imaging, 3D printing techniques, and intraoperative navigation to perform a partial facial and bilateral jaw graft in a patient who suffered facial ballistic trauma. The authors noted that the use of intraoperative navigation allowed for accurate fixation of allogeneic bone segments in the recipient. By using mobile intraoperative CT to perform a craniofacial CT scan after the formation of the recipient defect, it was possible to align and overlay the surgical plan on the patient's skeletal defect and implant the allograft under real-time intraoperative image guidance.
3.2 Orthognathic Surgery
Orthognathic surgery treats maxillofacial deformities caused by differences between the facial bones. This damage affects chewing, speaking, and breathing and may eventually even lead to tooth loss. Orthognathic surgery restores facial harmony as well as normal dental occlusion through skeletal cutting, repositioning, and fixation. However, in daily practice, orthognathic procedures have been faced with the limitations of traditional tools and a lack of intraoperative assistance. For the past several decades, dental molds simulating surgery and static navigation through manually fabricated splints have been routine practices in the treatment planning of patients with malocclusion. However, the limitations of this traditional method in the fabrication of dental membranes and splints also directly affect the accuracy of orthognathic surgery. Since the incorporation of 3D virtual planning and CAD/CAM technology into the orthognathic surgical planning workflow, traditional model surgery and splinting have been increasingly replaced by 3D virtual planning and CAD/CAM splint fabrication. Despite the improved accuracy compared to that in traditional surgery, the inherent shortcomings of static navigation surgical splints, such as mandibular rotation and a lack of vertical control of maxillary position, have not been well addressed (Van den Bempt et al. 2018).
After orthognathic surgical osteotomy, precise reproduction of the preoperative proximal mandibular position is difficult to achieve. Berger et al. (2018) performed a clinical pilot study to evaluate the accuracy of the electromagnetic navigation system for condylar positioning after high-oblique sagittal split osteotomy. Based on the experimental results, it was found that electromagnetic navigation condylar repositioning was equally accurate compared to manual repositioning. And electromagnetic navigation also prevented clinically hidden and severe condylar positioning errors (Fig. 5(a)). Lutz et al. (2019) developed a new tool for improved preoperative and postoperative simulation and intraoperative navigation in orthognathic surgery. They developed a semi-automated segmentation pipeline that allows the accurate and timely creation of patient-specific 3D models from CT scans for surgical planning. They have also developed software to simulate post-operative outcomes in facial soft tissue. Volumetric meshes were processed from segmented DICOM images using the Bullet open-source mechanical engine and mass-spring model to achieve postoperative simulation accuracy of <1mm. In addition, Lutz et al. developed a real-time navigation system using minimally invasive electromagnetic sensors. This navigation system features a new user-friendly interface based on enhanced virtualization, especially for trainee surgeons, to improve surgical accuracy and operative time, thus demonstrating its educational benefits (Fig. 5(b)). Lartizien et al. (2019) verified the benefit of intraoperative navigation in orthognathic surgery for learning condylar repositioning. A trainee performed repositioning of the condyle on one side of 100 patients undergoing bilateral sagittal splinting osteotomies in two stages. A comparative trial concluded that the navigation system is a promising original intraoperative learning tool for repositioning the condyle during bilateral sagittal splint osteotomy (Fig. 5(c)). Cao et al. (2020) evaluated the immediate improvement of orthognathic surgery performed in conjunction with conservative condylectomy based on intraoperative navigation with intraoral access and the long-term stability of postoperative results. The study included 56 patients with unilateral mandibular condylar osteochondroma combined with secondary facial asymmetry and malocclusion. All patients had successful surgery and healing, and the patients showed no signs of recurrence or temporomandibular joint ankylosis at follow-up. Facial symmetry was significantly improved in terms of chin deviation, chin rotation, and asymmetry indices of the three mandibular landmarks. Chen et al. (2021b) developed a novel surgical navigation system. The system can track the surgical saw in real time, and it can navigate the loose bone graft to align with the preoperatively planned position guided by interactive 2D and 3D views. Model experiments yielded an average error of 1.03±0.10mm for image-guided repositioning, which was significantly better than the average error based on non-navigation methods, validating the feasibility of this surgical navigation system. Han et al. (2021) evaluated the feasibility and accuracy of a robotic arm combined with intraoperative image navigation in orthognathic surgery. The experiments were performed on 12 intact cranial models. After the Le Fort I osteotomy, the maxilla was repositioned to different target positions and stabilized using the robotic arm and image navigation. The obtained maxillary position was compared with the planned position using navigation and postoperative CT images. Although the maxilla exhibited slight displacement during fixation, the mean absolute deviations from the target position in the medio-lateral, anterior-posterior, and superior-inferior directions were 0.16mm, 0.18mm, and 0.20mm, respectively. The mean absolute deviation of the obtained maxillary position was less than 0.5mm in all dimensions, and the mean root mean square deviation was 0.79mm compared to the target position using postoperative CT. Han et al. concluded that the robotic arm combined with intraoperative image-guided navigation has great potential for accurate repositioning of the maxilla during orthognathic surgery. Chen et al. (2021a) examined the accuracy of computer-assisted intraoperative navigation in bimaxillary orthognathic surgery on 45 patients with congenital dentomaxillofacial deformities who were scheduled to undergo bimaxillary orthognathic surgery. Intraoperatively, a Le-Fort I osteotomy of the maxilla was performed using an osteotomy guide. After the Le-Fort I osteotomy and bilateral sagittal split ramus osteotomy of the mandible, an occlusal splint was used to fix the mobile maxillary and distal mandibular segments to form the maxillary and mandibular complex. Real-time computer-assisted intraoperative navigation is used to guide the maxillary and mandibular complexes in the indicated directions. The actual clinical examination yielded a mean linear difference of 0.79mm (maxilla: 0.62mm, mandible: 0.88mm) and an overall mean angular difference of 1.20°. This study demonstrates the role of computer-assisted intraoperative navigation in the precise positioning of bone segments during bimaxillary orthognathic surgery and that intraoperative navigation was used as a reliable method to accurately transfer surgical plans during surgery (Fig. 5(d)). Yuan et al. (2022) evaluated the accuracy of AR surgical navigation combined with a pre-bent distractor in vector transfer by comparing it to a pre-bent distractor alone. They fabricated three identical 3D-printed cranial models for each patient based on preoperative CT data from 10 patients with maxillary dysplasia. One model was used for preoperative planning, while the other two models were used for experimental surgery in the AR+ pre-bent distractor and pre-bent distractor groups, respectively. The experimental results showed that the angular deviations of the traction vectors in space, the x-y plane, and the y-z plane were significantly smaller in the AR+ pre-bent distractor group than in the pre-bent distractor group, while there was no significant difference in the x-z plane. The AR+ pre-bent distractor group was more accurate in the deviations of the Euclidean distance and y-axis. The retraction results were more accurate in the AR+ pre-bent distractor group (Fig. 5(e)).
Fibrous dysplasia is the replacement of normal bone tissue by fibrous bone connective tissue and is a common benign craniofacial bone lesion. The primary goals of treatment for craniofacial fibrous dysplasia are to restore aesthetics and correct or prevent functional problems. Traditional surgical approaches make it difficult to accurately calculate the extent and amount of bone that needs to be reduced during surgery. CBCT has been shown to have comparable quality to MSCT in high-contrast anatomical structures such as bone. Generating 3D reconstructions by intraoperative CBCT has clear advantages in maxillofacial surgery and helps to minimize intraoperative and postoperative complications (Assouline et al. 2021). Goguet et al. (2019) applied mobile CBCT as an intraoperative navigation tool to perform an orthodontic procedure for left orbital-frontal fibrous dysplasia and a temporomandibular ankylosed joint block resection. The postoperative examination showed that the mobile CBCT device could effectively improve surgical accuracy and safety. However, they also pointed out that intraoperative CBCT increased radiation exposure and operative time (Fig. 5(f)). Liu et al. (2020) divided 24 patients with zygomaticomandibular fibrous dysplasia into two groups: patients who underwent surgery using virtual planning and surgical navigation were divided into Group A, and patients who underwent surgery through intraoperative assessment by the surgeon were divided into Group B. This was used to assess and compare the three-dimensional accuracy of osteomodeling procedures for zygomaticomandibular fibrous dysplasia performed by both surgical modalities. Based on actual clinical results, the mean root mean square was significantly lower in Group A than in Group B. Patients in Group A were significantly more satisfied with their facial symmetry. The mean operative time was comparable in both groups. Surgical technique and surgical area were important factors affecting the 3D accuracy of the surgery (Fig. 5(g)). Bao et al. (2020) quantitatively evaluated the effectiveness of surgical navigation in correcting zygomatic asymmetry in 26 patients with unilateral zygomatic fibrous dysplasia who underwent bone reconstruction. Patients were divided into two groups based on the presence or absence of intraoperative navigation, and all patients in both groups recovered successfully with improved facial symmetry and aesthetics, with better facial symmetry in the navigation group than in the conventional group and better postoperative curvature in the navigation group. This suggests that the use of intraoperative navigation can effectively improve the symmetry of the corrected zygoma (Fig. 5(h)). Gao et al. (2022b) applied head-mounted display-based AR navigation technology to the treatment of craniofacial fibrous dysplasia. The AR navigation technique was used to reconstruct the normal anatomical contour of the deformed area by superimposing the unaffected side onto the affected side through preoperative planning and 3D simulation. Intraoperative AR-guided facial bone reconstruction was performed successfully in all cases. The mean difference between the actual surgical repositioning and the preoperative plan was 1.036±0.081mm. The mean operative time was 66.4 minutes. The average preoperative preparation time was 29.6 minutes. The experimental results suggest that AR navigation-assisted facial bone reconstruction is a valuable treatment modality for craniomaxillofacial fibrous dysplasia, showing advantages in improving the efficiency and safety of this complex procedure (Fig. 5(i)).
3.3 Implant Placement
Currently, most dental implants are placed unassisted, without any form of computerized 3D planning. Surgeons perform osteotomies and place implants with their bare hands, using only adjacent and opposing teeth as a positional reference. In recent years, there have been tremendous developments in the clinical application of computer-aided systems due to rapid increases in computer computing power and improvements in imaging modalities. Among these, the use of surgical templates as CASN has allowed a high level of accuracy and is considered a reliable method to guide the placement of conventional implants in partially edentulous patients, but the lack of surgical experience may affect the accuracy of the procedure when using static guided navigation for implant placement. In contrast to CASN using surgical templates, CADIN uses high-precision motion tracking technology to track the position of the drill and patient throughout the procedure. This allows the surgeon to monitor the bone drilling and implant placement in real time during the entire procedure and to make real-time dynamic adjustments to the surgical plan based on the actual intraoperative situation (Gargallo-Albiol et al. 2019; Panchal et al. 2019; Wu et al. 2019; Block 2023).
Jiang et al. (2018) proposed a 3D AR navigation method based on point cloud image-patient alignment and evaluated its feasibility. The method allows merging virtual images for dental implantation in a real environment using a 3D image overlay. In model experiments, the root mean square deviation of the alignment was 0.54 mm, and the implant surgery results showed an average linear deviation of <1.5mm and an angular deviation of <5.5°. In the apical areas of the central incisor and the canine region, AR-guided implant surgery demonstrated small horizontal, vertical, and angular errors. The operative time using the AR navigation method was significantly shorter than that using the 2D image navigation method. In addition, volunteer experiments showed that the preoperative in situ 3D model could be accurately superimposed on the surgical site. The experimental results show that the method can achieve good alignment accuracy (Fig. 6(a, b)). Qin et al. (2019) developed and demonstrated an oral and maxillofacial navigation system called BeiDou-SNS with automatic fiducial point recognition. In all cases, the coverage of the Euclidean distance between the automatically selected fiducial marker positions and the manually selected fiducial marker positions by an experienced dentist ranged from 0.373mm to 0.847mm. Under the guidance of BeiDou-SNS, four implants were placed in the 3D-printed model. While the maximum deviations between the actual and planned implant entry and end points were 1.328mm and 2.326mm, respectively, the angular deviations ranged from 1.094° to 2.395°. The results show that the oral surgery navigation system with automatic recognition of fiducial marker positions can meet the requirements of clinical surgery. Aydemir and Arisan (2020) compared the deviation of planned and placed implants with the assistance of a micron tracker-based dynamic navigation device and freehand methods. Navigation-guided implant surgery provided a maximum deviation of only 0.7mm linearly and a 5° axial angle. Experimental results showed that the use of a dynamic intraoperative navigation system greatly increased the accuracy of implant placement (Fig. 6(c)). Ochandiano et al. (2021) proposed a new workflow for dental implant restorations. Virtual dentures were designed by scanning CT data and jaw plaster models. The virtual was then planned and transferred to the patient for final intraoperative validation by a combination of intraoperative infrared optics navigation and 3D printed guidance brackets supported by conventional static teeth, intraoperative dynamic navigation, and AR. It has been demonstrated experimentally that there is a significant learning curve for the surgeon when applying the guidance method. Initial navigation cases achieved lower accuracy due to instability in fixture alignment but were comparable to non-navigation freehand positioning. Subsequent dynamic navigation cases incorporating highly stable static guides as reference and alignment markers yielded the highest accuracy at the insertion point with a deviation of 1-1.5mm. Ochandiano et al. also noted that smartphone-based AR visualization is a valuable tool for intraoperative visualization and final validation (Fig. 6(d, e, f)). Järvinen et al. (2021) used virtual surgical planning combined with intraoperative navigation to place patient-specific implants in 5 patients who underwent bilateral sagittal split mandibular osteotomies and in 5 patients who underwent mandibular osteotomies. All patient-specific implants were precisely placed in the mandible, and the screws were inserted into the pre-drilled screw holes, obtaining a predetermined occlusion. At the same time, however, Järvinen et al. noted that although the intraoperative navigation system offered the possibility to verify the surgical results during the procedure, the accuracy was not sufficient to be used as a virtual drilling guide alone.
Zygomatic implants are ideal for the treatment of severely atrophic edentulous maxillae and maxillary defects, as they replace the widely used bone augmentation method and shorten the treatment period. Wang et al. (2018) applied a commercial navigation system for zygomatic implantation in 15 patients with severe maxillary atrophy. Postoperative results showed that all zygomatic implants were fixed in the maxillary alveolar process and the zygomatic bone, and postoperative radiological examination showed that no critical anatomical structures were damaged during the implantation process. The overall survival rate of all zygomatic implants that achieved osseointegration was 100% after early healing. Shen et al. (2021) described a practical, feasible, and reproducible protocol for a real-time surgical navigation system for the precise placement of four zygomatic implants in patients with severe atrophy in the maxilla whose residual bone does not meet the requirements for conventional implants. Hundreds of patients treated with zygomatic implants have been operated on using this protocol. Clinical results have been satisfactory, with low intraoperative and postoperative complications. Applying this approach throughout the procedure ensures a safe zygomatic implant placement procedure (Fig. 6(g)).
Alveolar cleft bone grafting is now a commonly used surgical procedure to repair alveolar bone defects in patients with cleft lip and palate. Zhang et al. (2021b) successfully operated on three patients with alveolar clefts under navigational guidance. With the model alignment procedure, an exact intraoperative match was achieved between the actual position and the CT images with a systematic error within 0.3mm. With the help of the instruments and probe navigation, the grafted bone was accurately placed according to the preoperative plan. All patients healed well with no serious complications. The results of the study demonstrate that dynamic navigation is a valuable option for handling the complex surgical procedure of alveolar fissure repair.
3.4 Maxillomandibular Surgery
In CADIN, precise registration is critical because it directly affects the accuracy of all subsequent navigation tasks. In intraoperative navigation for maxillary surgery, the current common alignment method is to use a positioning screw implanted into the maxillary alveolar bone to locate the maxilla, and while this method provides accurate positioning, it can also undoubtedly be invasive to the patient. Yamamoto et al. (2020a) integrated a novel splint developed with a reference frame into maxillary navigation surgery using a marker-based point alignment procedure. Using this customized integrated splint, the marker-based point alignment procedure can be done easily and noninvasively, which can assist the surgeon in maxillary surgery with high precision. In addition, the splint does not require re-registration during surgery and can be temporarily removed when needed. This approach also provides the greatest convenience to the surgeon (Fig. 7(a)).
Visualization of complex anatomical structures and safe and effective osteotomy solutions are the basis for CADIN systems to accurately guide craniomaxillofacial surgery. For mandibular surgery, the mobile nature of the mandible makes the registration procedure for intraoperative navigation complex, resulting in a significant limitation in the application of CADIN techniques in mandibular surgery. To address the problem of alignment accuracy for mandibular surgery navigation, Yamamoto et al. (2020b) developed a new alignment strategy for mandibular navigation surgery using an open splint that combines a reference frame and alignment markers. The alignment strategy can be clinically shown to be easy and noninvasive to perform marker-based alignment procedures and to perform mandibular navigation procedures with high accuracy. In addition, the splint can be removed when needed and re-registered very easily. However, the authors also suggest that although this strategy cannot be applied to edentulous patients or patients with only a few remaining teeth, it may be suitable for many types of mandibular navigation procedures, such as for the protection of the IAN bundle during osteotomy procedures and for applications during the removal of large benign tumors, removal of foreign bodies, and implantation of teeth in the mandible (Fig. 7(b)). Ma et al. (2019) developed an intelligent surgical system that integrates a navigation system and a robot to reduce the impact of human factors on oral and maxillofacial surgeries. They conducted drilling experiments on five 3D-printed mandibular models to test their positional detection capability and evaluate their operational performance. The experimental results showed that the system was able to successfully guide the robot through the surgery regardless of mandibular posture. Both software and hardware accuracy were acceptable. Brouwer de Koning et al. (2021) evaluated the feasibility of electromagnetic navigation for osteotomy guidance in patients undergoing surgery for mandibular tumors. Intraoperatively, the position of the mandible was tracked using an electromagnetic sensor fixed to the mandible. The real-time positions of the mandible and a pointer were displayed on the navigation system. The experiment was performed on 11 patients. The target registration error values were 3.2±1.1mm and 2.6±1.5mm using anatomical markers and markers on the cutting guide, respectively. The navigation procedure added an average of half an hour to the operative time (Fig. 7(c)). Hou et al. (2022) developed a novel occlusal splint-based optical navigation system for craniomaxillofacial surgery. They selected 10 Beagle dogs to undergo bilateral mandibular osteotomies under the occlusal splint-based navigation system. The average positional deviations between the preoperative design and intraoperative navigation were experimentally derived as 0.01±0.73mm on the lateral height of the mandibular ramus, 0.26±0.57mm on the inner height of the mandibular ramus, and 0.20±0.51mm on the osteotomy length. The average deviation of the angle between the mandibular osteotomy plane and ramus plane and the mandibular branch plane was 0.94°±1.38°, and the average error on the angle of the retained mandibular angle was 0.66°±0.97°, and the experimental data showed that the system had good consistency (Fig. 7(d)).
AR is a technology that calculates the position and angle of camera images in real time and adds corresponding images, videos, and 3D models. The goal of this technology is to put the virtual world on the screen and interact with the real world. In recent years, AR systems have gradually started to be used in oral and maxillofacial surgery because of their ability to provide physicians with sufficiently clear, intuitive medical images and to provide accurate image navigation for practical applications (Ayoub and Pulijala 2019; Jiang et al. 2023; Jiang et al. 2022). Zhu et al. (2018) validated the accuracy of AR navigation in mandibular osteotomy by comparing it with other interventional techniques in a retrospective study of 93 patients undergoing surgery for mandibular hypertrophy. There was no significant difference in overall operative time between the three groups compared to the AR group and the group using the personalized template, although the operative preparation time was much shorter in the freehand group. In addition, the difference in osteotomy line between the AR group and the group using personalized templates was smaller than that of the freehand group. The experimental results suggest that the AR-based navigation system has a good clinical prospect with higher accuracy, higher reliability, and better user friendliness in some specific clinical procedures compared with other techniques (Fig. 7(e)). Pietruski et al. (2019b) presented a novel AR-based navigation system that provides good functionality and ergonomics. The results of a simulated osteotomy performed using this navigation system showed that projecting the virtual surgical plan into the surgeon's field of vision improved hand-eye coordination and positioning within the surgical area. As a result, the results of the surgery were similar to those obtained using the cutting guide. Han et al. (2022a) constructed a naked-eye 3D display using an eye-tracking module and an autostereoscopic display device to track the position of the prototype using an optical tracking system for real-time in situ surgical navigation. They performed a maxillary drilling experiment and obtained an average position error of 1.12±0.30mm and a positioning time of 8.7±3.9s. The experimental results show that the system can provide high-quality AR images for surgical navigation, thus reducing positioning errors and operation time (Fig. 7(f)).
The temporomandibular joint is the only left and right bilaterally linked joint in the maxillofacial region and has a certain degree of stability and multi-directional mobility. It produces various important activities related to mastication, swallowing, speech, and expression under the action of muscles. Temporomandibular joint ankylosis not only leads to functional deficits but also manifests as maxillofacial deformities. Facial nerve injury and the close proximity of the middle meningeal artery to the neck of the condyle are the two main obstacles to temporomandibular joint surgery. Newman et al. (2018) established a protocol based on the utilization of intermaxillary fixation screws to produce accurate and reproducible results using intraoperative navigation during total joint replacement for the correction of bilateral temporomandibular joint ankylosis. They noted that intraoperative navigation not only allows for real-time identification of important structures but also helps confirm that no rotation or movement of the mandibular complex is occurring, which is particularly beneficial for the use of custom joints. In addition, navigation can be used to verify the field preparation of the custom joint prosthesis (Fig. 7(g, h)). Chowdhury et al. (2020) performed temporomandibular joint ankylosis surgery under navigational guidance. After obtaining CT images of the head and neck, the patient's artery and vein, which are in close proximity to the mandibular condyle, were marked using special software from Brain Lab. During the surgery, using the Brain Lab navigation system, the location of the middle meningeal artery was determined, and an osteotomy cut was performed. Neuhaus et al. (2021) investigated the feasibility and benefits of real-time navigation and intraoperative 3D imaging for use during total temporomandibular arthroplasty, as well as the clinical outcomes of patients. 26 custom-made prostheses were implanted in 21 patients by setting up a control group. Data from clinical trials suggests that real-time navigation can aid in positioning during lateral skull base dissection and resection. However, the results of real-time navigation-assisted drilling were not accurate in terms of vector and length control (Fig. 7(i)). Gomez et al. (2021) presented two cases of orthognathic surgery with computer-assisted surgical guidance for unilateral and bilateral temporomandibular replacements. In the first case, the difference in millimeters between planning and surgical outcomes was 1.72mm for the glenoid component and 2.16mm for the condylar prosthesis; for the second case, differences in the right side were 2.59mm for the glenoid component and 2.06mm for the ramus, and in the left side, due to the anatomy the difference was a little greater, without clinical significance. Based on clinical results, they concluded that combined midface mandibular surgery with temporomandibular replacement is feasible. Computer-assisted surgery facilitated the planning and design of the custom prosthesis and the execution of the procedure.
3.5 Removal of Foreign Body
The head and neck are the most vascularized areas, making it easy for foreign bodies to cause damage to vital organs, leading to serious life-threatening consequences such as hemorrhage, airway obstruction, and asphyxia. In most surgeries requiring foreign body removal from the craniomaxillofacial region, the normal anatomy of the maxillofacial region is altered or damaged, and accurate identification of the location of the foreign body is critical to the surgical procedure due to its proximity to vital structures and the difficulty of access. Although preoperative CT scans and 3D image reconstruction can help precisely locate the object and provide a clear image of the surrounding anatomy, even if the surgeon knows the exact location of the foreign body preoperatively, it is difficult to detect it accurately and quickly intraoperatively. Because the CADIN system can provide the surgeon with the location of the foreign body in 3D space in real time intraoperatively, it makes the CADIN system very effective in aiding surgery for the removal of oral and maxillofacial foreign bodies, which can greatly reduce the operation time (Sukegawa et al. 2018; Ji et al. 2018).
Ji et al. (2018) successfully debrided several glass fragments close to the internal carotid artery and parotid gland with the help of the AccuNavi-A surgical navigation system. Postoperative CT showed that all foreign bodies were removed, and the patient had no complications during the subsequent treatment. Zhang et al. (2020) proposed two solutions to optimize surgical navigation procedures: the use of a 3D-printed custom mandibular fixator to indirectly maintain consistency in preoperative planning and intraoperative navigation of the visual image of the foreign body. As well as introducing real-time endoscopic imaging during surgery to provide visualization under complex anatomical structures. With the aid of optimized surgical navigation, all patients had successful minimally invasive removals of the foreign body. No complications were identified at the 3- to 6-month postoperative follow-up. Based on actual clinical results, it has been shown that improved navigation accuracy and the provision of true vision can effectively compensate for the lack of navigation accuracy due to navigation positioning errors or interference from difficult-to-detect complex anatomical structures.
3.6 Skull Base Surgery
Skull base surgery has long been a challenging area of anatomy. Multiplanar reconstruction with MRI or CT imaging combined with virtual 3D planning is now increasingly used in skull base surgery to achieve accurate preoperative planning and to assist the surgeon in the actual clinic with CADIN to remove or sample lesions within safe limits that are difficult to reach without extensive surgical exposure (Rothweiler et al. 2021).
Tang et al. (2021) retrospectively analyzed data from seven patients who used a combination of mixed reality and surgical navigation to assist in oral and maxillofacial tumor surgery and evaluated the feasibility and accuracy of the protocol. Of the 7 patients, 4 had maxillary tumors and 3 had mandibular tumors. There were 13 osteotomy planes. The mean deviation between the planned and actual osteotomy planes was 1.68±0.92mm; chromatographic analysis showed an error of ≤3mmat 80.16% of the points. The mean deviation of the maxillary and mandibular osteotomy lines was approximately 1.60±0.93mm and 1.86±0.93mm. The mean follow-up was 15.7 months, and no tumor recurrence or complications occurred. The experimental results suggest that mixed reality technology combined with surgical navigation is feasible, safe, and effective in oral and maxillofacial tumor resection (Fig. 8(a)). Dean et al. (2022) successfully applied computer-assisted navigation piezoelectric surgery in a case of resection of an intraosseous hemangioma on the right supraorbital rim and in a case of resection of a primary T4a adenoid cystic maxillary carcinoma. During the surgery of the tumor patient, the surgeon was able to examine the bone surface and palatal mucosa as well as the tumor resection margins of the deep internal structures of the bone according to a preoperative virtual plan, and the surgery could be performed according to the surgical plan with an accuracy of 1 mm (Fig. 8(b)). Tel et al. (2022) provided an innovative approach for the removal of intraorbital masses based on multimodal image fusion, segmentation, virtual modeling, digital planning, and navigation. The procedures performed using the proposed method in 19 open and endoscopic procedures were successful. No intraoperative complications were reported. Actual clinical results showed a mean difference of <1mm for lateral osteotomies and <0.5mm for endoscopic osteotomies. Implementation of virtual planning and intraoperative navigation of the surgical procedure for removal of orbital masses enhances preoperative insight into orbital spatial anatomical relationships and improves the accuracy of the open and endoscopic approaches (Fig. 8(c)). Chan et al. (2022) used an AR system for intraoperative navigation in order to improve the incisional margins of maxillary carcinoma. The AR system can depict the tumor margins while addressing the deficiency of gaze switching for intraoperative navigation. Preoperatively, osteotomy planning is performed, and the AR system is used to project this information into the surgical field. Based on the results of model experiments, AR osteotomies were found to be highly similar to pre-planning. The AR-guided simulations had higher workload scores in the areas of psychological needs, performance, effort, and frustration. The projector-based AR approach improved tumor edge depiction (Fig. 8(d, e)).
Adenoid cystic carcinoma is a rare and aggressive tumor for which total resection with clear margins is the main treatment. However, the limited line of sight in this location and the high risk of injury make the procedure challenging. García-Sevilla et al. (2021) evaluated the accuracy of three types of less invasive navigation using optical tracking, 3D printing, and AR in a patient-specific model, all obtaining an error of less than 1 mm. Navigation software was used to guide tumor resection in a clinical case. Points were collected along the surgical margins after resection and compared to the real points identified in postoperative CT. Distances of less than 2mm were obtained in 90% of the samples. Postoperative CT scans showed adequate resection margins and confirmed that the patients were disease-free after two years of follow-up. García-Sevilla et al. noted that navigation provides surgeons with confidence that they can take a less invasive and more conservative approach to surgery (Fig. 8(f)).
Preoperative pathological diagnosis is a key step in the diagnosis of skull base lesions, and obtaining a sufficient number of specimens through a biopsy of the lesion can provide accurate lesion information for subsequent treatment. However, because of the deep location of the skull base area, the biopsy procedure is difficult to access and can easily cause damage to the structures surrounding the lesion. Currently, ultrasound guidance is often used in lesion biopsy procedures, but ultrasound itself is poor at detecting bone tissue as well as neurovascular tissue, which is not feasible in the skull base region with complex structures. Zhu et al. (2021) evaluated the diagnostic accuracy of navigation-guided core needle puncture biopsy for skull base and parapharyngeal lesions. They used an optical navigation system for preoperative design and intraoperative real-time image guidance in 20 patients with skull base and parapharyngeal lesions. Specimens were collected using a spring-loaded semiautomatic biopsy gun and biopsy needle. All patients underwent successful puncture biopsies without any immediate or delayed complications, with a diagnostic accuracy of 90%. They concluded that navigation-guided core-needle biopsy provides a simple method for the diagnosis of skull base and parapharyngeal lesions with high specimen yield and good patient tolerance (Fig. 8(g)). Yang et al. (2021) performed biopsies on 8 patients with infratemporal-middle cranial fossa communicative tumors using navigation-guided CT-MRI image-based fusion and analyzed the clinical data of postoperative patients to assess the outcome of treatment under surgical navigation. In four of these cases, a definitive pathological diagnosis was obtained from the core needle biopsies under navigation guidance. Complete resection of the tumor was achieved in six of the seven navigation-guided procedures. The experimental results showed that the combination of CT-MRI image fusion and computer-assisted navigation management optimized the accuracy, safety, and surgical outcome of core-needle biopsy and surgery for infratemporal-middle cranial fossa communicative tumors (Fig. 8(h)).
Accurate determination of lymph node status is crucial for cancer staging and adjuvant therapy. A case in point is oral squamous cell carcinoma. Chand et al. (2018) describe an innovative approach to anterior sentinel lymph node guidance that uses radionuclide tracers, 3D AR-guided imaging, near-infrared fluorescence overlay imaging, and handheld probes to optimize accuracy, efficiency, and intraoperative navigation of anterior lymph node localization in head and neck cancers. In a case of cT1N0M0 squamous cell carcinoma of the tongue, preoperative radionuclide lymphatic scintigraphy was performed with a sentinel lymph node-specific radiolabeled tracer. Intraoperatively, a 3D handheld AR scanning SPECT probe was used to assess the consistency of the anterior sentinel lymphatic sites with the preoperative SPECT-CT images. Real-time optical video was linked to the SPECT-CT images to improve accuracy. Final guidance of the anterior lymphatic sites was performed using ICG fluoroscopic imaging. Dynamic navigation and SPECT-CT showed bilateral lymphatic drainage of the tumor. They concluded that the combination of AR, nuclear medicine, and superimposed fluorescence imaging allowed for more accurate matching of preoperative imaging with intraoperative identification and precise guidance of dissection.
3.7 Endodontic Treatment
Root canal preparation is the first stage of endodontic treatment. This critical step must allow for accurate localization of the root canal and its cleaning and shaping to maximize the elimination of microorganisms. Inaccurate root canal preparation procedures can lead to complications such as root canal omissions, root canal perforation, fracture, and coronal weakening, which can seriously affect the survival of the tooth. Pulp canal obliteration or calcified canals, is the narrowing of the pulp chamber due to the deposition of hard tissue in the root canal space. Teeth with pulp canal obliteration or calcified canals usually require challenging and time-consuming endodontic treatment (Ribeiro et al. 2022) . When performing endodontic treatment in patients with pulp canal occlusion, despite the application of imaging techniques such as CBCT, access cavity preparation is prone to procedural errors that may result in substantial loss of dentin structure, thus reducing long-term prognostic outcomes.
Lara-Mendes et al. (2018) described a technique for endodontic treatment through a new minimally invasive approach using CBCT imaging and 3D guidance without damaging the incisal edge of the tooth. After radiographic examination of a patient who had suffered dental trauma, they found that the patient had a slight thickening of the apical periodontal ligament space, and the visible canal space was limited to the apical 2mm portion of the tooth root. After an intraoral scan of the tooth surface, guided root canal treatment was planned with a CBCT scan. With the help of virtual implant software, a virtual model was created for surgical access planning in a manner that did not damage the incisal edge of the tooth. A follow-up was performed one year after the completion of treatment. The patient was asymptomatic, and the periapical tissue was within normal limits. Despite the presence of a severely calcified root canal, guided root canal treatment optimized the treatment outcome, providing conservative access in a safe and predictable manner without causing tooth damage (Fig. 9(a)). Chong et al. (2019) investigated a new use of computer-aided dynamic navigation in guided endodontic treatment. They planned the entry point, angle, path, and depth of the drill hole for virtual implants for 29 selected teeth. Conservative access cavities were obtained for 26 of these teeth, and all expected root canals were successfully located. Due to tracking difficulties, only one root canal was found in two maxillary second molars; in one maxillary first molar, only two root canals were found, and the third root canal was off target with a misaligned entry preparation. The results of the study demonstrate the potential of using computer-assisted dynamic navigation techniques for endodontic guidance in clinical practice. The combination of CBCT imaging and intraoral scanning of the target area can produce access navigation that is very helpful in locating severely calcified root canals in complex cases (Fig. 9(b)). Jain et al. (2020a) proposed a new dynamic navigation method for achieving minimally invasive access preparation and performed experiments on 3D printed jaw models to evaluate its accuracy in locating highly calcified root canals. Digital measurements of 2D and 3D horizontal, vertical, and angular differences between the planned and actual access preparation were performed using superimposed CBCT scans, showing a mean 2D horizontal deviation of 0.9mm, with significantly higher deviations in the maxillary teeth than in the mandibular teeth. The mean three-dimensional deviation from the canal opening was 1.3mm, which was slightly higher in the maxillary teeth compared to the mandibular teeth. The mean three-dimensional angular deviation was 1.7°, which was significantly higher for the molars compared to the premolars. The average drilling time was 57.8s. This study shows the potential of combining dynamic 3D navigation techniques with high-speed drills to preserve tooth structure and accurately locate root canals in patients with pulp canal occlusion. In addition, Jain et al. (2020b) compared the speed, quality accuracy, and quantitative loss of tooth structure in 3D printed teeth with simulated calcified root canals using a freehand unguided operation and dynamic navigation guidance for root canal localization. Results from experiments performed randomly on 40 maxillary and mandibular central incisors under simulated clinical conditions yielded significantly less average material loss with dynamic navigation access compared to the freehand technique. Dynamic navigation was also more accurate in locating calcified pulpal approaches. Substance loss in mandibular teeth was negligible between the two access techniques. Overall, root canal preparation was significantly faster with the dynamic navigation technique than with freehand preparation. Zubizarreta-Macho et al. (2020) analyzed the accuracy of computer-assisted dynamic navigation and static navigation techniques compared to freehand in guiding root canal preparation. By dividing 30 teeth into three groups for the experiment, it was concluded that there was no statistically significant difference between static and dynamic navigation at the coronal, apical, or angular levels; however, there was a statistically significant difference between the two computer-assisted navigation techniques and the conventional access technique at the coronal, apical, and angular levels. The experimental results showed that both computer-assisted static and dynamic navigation procedures can accurately perform endodontic access (Fig. 9(e, f)). Dianat et al. (2020) compared the accuracy and efficiency of a dynamic navigation system and a freehand, unguided method for locating calcified root canals. After experiments with 60 teeth suffering from pulp canal occlusion, it was found that the dynamic navigation system group had significantly lower mean linear and angular deviations, reduced dentin thickness, time required for preparation to enter the pulp chamber, and number of accidents than the freehand group. The experimental results suggest that the dynamic navigation system is more accurate and effective than the freehand technique in locating the calcified canals of human teeth. Torres et al. (2021) evaluated the 3D accuracy and outcome of dynamic navigation methods for guided endodontic treatment of teeth suffering from severe pulp canal occlusion in 3D-printed jaws. 156 root canals were detected by 3 doctors in pairs of 168 affected teeth, with an overall success rate of 93% and no difference in operator experience. The mean deviation from the apical point was 0.63mm, which was significantly lower in the anterior teeth compared to the molars. The mean angular deviation from plan was 2.81°. The experimental results suggest that dynamic navigation is an accurate method for treating severely calcified root canals. Torres et al. also noted that the technique has a learning curve and requires training prior to clinical use (Fig. 9(d)).
A common imaging tool for guided endodontic treatment is CBCT combined with intraoral scan images for virtual access planning. However, CTCB scans increase the patient's radiation exposure of the patient. Leontiev et al. (2021) verified the accuracy of MRI for guided endodontic detection of root canals. It was experimentally validated that 91 out of 100 root canals were successfully explored with MRI-guided root canal preparation. The average angular deviation was 1.82°. The average deviation was 0.21-0.31mm at the base of the drill and 0.28-0.44mm at the tip of the drill. This in vitro study demonstrated the suitability of MRI for guiding the preparation of the pulp access cavity.
Removing the fiber posts from an endodontically treated tooth is often necessary, especially for teeth with failed root canal therapy. Removal of the fiber posts from the root canal is also risky, and removal of the fiber posts is often a challenge for the surgeon. To mitigate the risks of the procedure, procedural errors need to be avoided. Janabi et al. (2021) compared the accuracy and efficiency of removing fiber posts from root canal-treated teeth using a 3D dynamic navigation system and a freehand unguided operation. A comparative trial of 26 maxillary teeth revealed that 3D dynamic navigation showed less deviation and angular excursions in the overall coronal and apical planes, and the operation required less operative time compared to the freehand operation. In addition, the volume loss to the tooth structure was significantly less with the 3D dynamic navigation technique than with the freehand technique. The results of the trial showed that 3D dynamic navigation was more accurate and efficient in extracting fiber posts from root canal-treated teeth compared to freehand manipulation (Fig. 9(c)).
The preservation of important anatomical structures is necessary when performing root-end resection or periapical surgery. However, precise root end resection is difficult to achieve clinically due to a limited field of view, an inconvenient perspective, and intrusive bleeding (D et al. 2022). Han et al. (2022b) performed osteotomies and root end resections in 16 patients, guided by dynamic navigation. They concluded that the method was accurate for locating root tips located far from the labial/buccal cortical bone location. Dianat et al. (2021) verified the accuracy of dynamic intraoperative navigation in guiding root-end resection. The experimental results showed that linear deviation, angular deflection, and operative time were significantly reduced under the guidance of dynamic intraoperative navigation.